Enhanced deep learning networks integrated by fractals for air quality index analysis

The Air Quality Index (AQI) provides daily information on the quality of outdoor air and increases with rising air emissions. Accurate AQI analysis and prediction are important for understanding and managing air pollution. This study develops an integrated deep learning framework incorporating a fractal approach to analyze and predict AQI.

The developed framework consists of two main steps. First, AQI data are pre-processed using a fractal interpolation technique to address data discrepancies. Second, the pre-processed data are trained and tested using long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM), and convolutional neural network with long short-term memory (CNN-LSTM) models for AQI prediction.

The proposed models are evaluated using AQI data from five major cities in India. Statistical performance metrics are used to assess and compare the predictive performance of the developed models.

The integration of fractal interpolation with deep learning provides a framework for handling discrepancies in AQI data and improving the analysis and prediction of air quality. The developed fractal-integrated deep learning models demonstrate their applicability for AQI prediction across the selected major Indian cities.